3D silicon detectors 1. Introduction The LHC saw a change in the

October 14, 2011 14:51 WSPC - Proceedings Trim Size: 9in x 6in BatesR˙Como˙3Dsensors˙Paper

1

3D silicon detectors

R. L. Bates∗

SUPA School of Physics and Astronomy, The University of GlasgowGlasgow, G12 8QQ, UK

∗E-mail: [email protected]://ppewww.physics.gla.ac.uk/∼batesr/

Significant process in 3D detectors has taken place since Sherwood parker pro-

posed the 3D silicon detector in 1997. The 3D detector was conceived as amethod to overcome the radiation induced reduction in carrier lifetime in heav-ily irradiated silicon detectors via the use of advanced MEMS device fabrica-

tion techniques. This paper reviews the state of the art in 3D detectors. Workperformed within the major fabrication institutes will be discussed, includingmodifications to the original design to reduce complexity and increase deviceyield. Characterization of 3D detectors up to the maximum radiation fluence

expected at the high luminosity LHC operation will be presented. Results fromboth strip and pixel devices will be shown using characterization methods thatinclude 90-Sr betas, focused laser and high-energy pions.

Keywords: silicon detector, 3D detector, radiation hard, pixel detector

1. Introduction

The LHC saw a change in the paradigm in the use of silicon detectors, in

particle physics, from small scale vertex detectors to their use as large area

trackers using both silicon strip and pixel detector assemblies. The predicted

radiation fluence of the LHC experiments drove the development of silicon

sensors that can be operable at fluence levels of 1015 cm−2 1MeV neq, which

was unprecedented at the time of planing the LHC experiments.

Over the next ten years the LHC based experiments will undergo up-

grades to extend their physics reach. For example, the ATLAS exper-

iment will place an additional pixel layer inside the existing pixel sub-

detector at a minimum radius of 31 mm from the interaction point, known

as the Inner B–layer1 (IBL). This will expose the silicon detectors to a

fluence of 3 x 1015 cm−2 1MeV neq. Including necessary safety factors,

the design requires the silicon detectors to be operable at a fluence of


2

5 x 1015 cm−2 1MeV neq. In addition the LHC will be upgraded, to the

high luminosity LHC (HL-LHC), to deliver a factor of ten increase in the

integrated luminosity. This will, in turn, increase the radiation field inside

the experiments by the same ratio. As a consequence the silicon pixel de-

tectors will have to survice a fluence up to 1016 cm−2 1MeV neq. Which

is a new radiation tolerance frontier. To this end new operational modes

and device structures are being investigated for silicon detectors. The 3D

detector is such a new device structure.

2. 3D silicon detector design

The 3D detector was first suggested by S. Parker2 as a method to over-

come the radiation induced reduction in the charge carrier mean free

lifetime3,4 which becomes a problem after a radiation fluence of order

1015 cm−2 1MeV neq. At such a high radiation level full charge collection

in planar silicon detector is not expected.

The 3D detector has an array of n- and p-type electrode columns passing

through the silicon substrate rather than being implanted on its surface, as

shown in Fig. 1. These electrodes are realised by a combination of deep re-

active ion etching to realise the holes and low pressure vapour deposition to

fill them with a dopant.5,6 Standard detector technologies are used for sur-

face structures. The design allows the combination of a standard substrate

thickness with a lateral electrode spacing of a few tens of micrometers.

Therefore the depletion and charge collection distances are reduced, with-

out reducing the sensitive thickness of the detector. This implies that the

device is extremely fast, has high charge collection and a low operating

voltage, and therefore low power consumption, even after a high irradiation

dose. The enclosed structure of the unit cell of the 3D detector will also

reduce the amount of charge sharing, which could be advantageous in in-

creasing the signal in a given pixel after heavy irradiation. These features

should make 3D detectors substantially more radiation hard than standard

planar devices.7

2.1. Full 3D sensors

The original 3D detector, known as the full 3D detector, is fabricated using

one sided processing. It requires the sensor wafer to be wafer bonded to

a support wafer and for the electrode holes to be completely filled with

doped polysilicon. Both electrode types are connected together on the top

side of the device as the fabrication is a single sided process. The back side


3

Fig. 1. Schematic of a 3D detector.

handeling wafer maybe removed to reduce material. The full 3D detector

allows the addition of an electrode around the full detector matrix that

contains the electric field and allows the detector to be active all the way

to the edge of the device, known as an active edge.8 Full 3D detectors have

been fabricated at both Sintef9,10 and Stanford.5

Sensors that are compatible with the ATLAS FE-I311 front-end pixel

readout chip have been fabricated at Sintef, assembled and tested at

CERN.12 They show good electrical characteristics and full charge collec-

tion. The carrier lifetime in the filled column should be sufficient to collect

significant charge from radiation incident upon the column. However, the

charge collected in this region has been observed, with a fined focused x-ray

beam, to be reduced from that expected. It is believed that the reason is

that during the fabrication process an oxide is formed on the edge of the

column which forms a barrier to carrier collection from the column.13 A

change in the dopant chemistry removing oxygen should reduce the trap-

ping and increase charge collection from inside the column.

2.2. Double-sided 3D sensors

An alternative fabrication process, know as the double-sided 3D detector,

was introduced6,14 to reduce 3D detector fabrication complexity. This uti-

lizes double-sided mask alignment technology which enables the columns of


4

one type to be etched from the opposite side of the wafer to the other type,

removing the need for a support wafer. The holes may pass all or part-way

through the sensor wafer. To reduce stress in the wafer the holes are not

completely filled with polysilicon and are therefore not active. The reduced

fabrication complexity has increased the yield of the 3D device. However if

the columns do not penetrate the full thickness of the wafer the sensor per-

formance is sensitive to the penetration depth of the columns which needs

to be well controlled. There is a lower field region in the detector directly

above a column which requires the detector to be over depleted to be fully

active.15

The double-sided 3D detector lacks the active edge, therefore to min-

imise the edge dead region the guard fence structure was developed.16 This

consists of multiple ohmic columns around the active silicon. The ohmic

columns stop the depletion region spreading from the pixel to the cut edge.

With a guard fence ohmic column spacing of 50 µm the cut edge can be

as close as 100 µm from the active silicon without adversely affecting the

device current-voltage characteristics.

3. 3D for CMS

Full 3D pixel sensors compatible with the CMS pixel readout chip have

been fabricated by Sintef.17 The pixel unit cell is 150 µm x 100 µm in

size. Two different device configurations have been investigated: four junc-

tion columns (4E) and two junction columns (2E) per pixel. The distances

between the centers of neighbouring junction and bias columns of the 2E

and 4E sensors are 62.5 µm and 45 µm, respectively, while the diameter of

the columns are 14 µm. The matrix is surrounded by an active edge. The

sensors have been integrated into assemblies and tested.18

The device breakdown takes place at 100 V, well above the full depletion

voltage of less than 40 V. The four electrode device has, as expected, higher

noise than the 2E device due to the higher capacitance load. The noise is

observed to fall at full depletion to between 250 and 300 electrons for the

2E configuration, compared to 100 electrons for a planar device.

From the preliminary beam test studies with 120 GeV protons, a signal

to noise ratio (S/N) of 80 and a signal to threshold ratio (S/T) of 3 have

been obtained for the non-irradiated 285 µm thick 2E detector. This S/T

value is too low for the successful operation in the CMS detector and more

work is being undertaken to lower the threshold value.


5

4. ATLAS IBL campaign

The double-sided 3D sensor has been recommended to be used for either

25% or 50% of the sensors in the ATLAS IBL. Both FBK and CNM are

processing the sensors using a mask set that is as common as possible. Pixel

sensors compatible with the ATLAS FE-I419 front-end pixel amplifier chip

have been fabricated and extensively tested both before and after irradia-

tion.20,21 The sensor show a good uniformity over the matrix for threshold,

noise and detection response. A noise performance of 150 electrons at a

threshold of 3200 electrons have been routinely obtained before irradiation.

After an irradiation of 5 x 1015 cm−2 1 MeV neq from reactor neutrons

the module threshold and noise performance is only slightly altered, with a

noise increase of only 5 electrons to 155 electrons, when measured at -15◦C.

The current of the device increased as expected.

Test beam results show good detection efficiency and charge collection

efficiency over the matrix before irradiation with and without a magnet

field.20 After irradiation to 6 x 1015 cm−2 1 MeV neq 3D and planar sensors

coupled to the FE-I4 pixel chip show similar hit detection efficiencies of

approximately 97%. For tracks normal to the device low efficiency regions

are observed around the electrodes. These appear as smeared regions of

reduced efficiency for non-normal track angles.22

5. 3D detector test beam analysis

Many test beams have been performed on 3D detector modules. This paper

concentrates on a test beam performed in a 120 GeV pion beam with CNM

double-sided 3D pixel detectors coupled to the TimePix23 readout chip,

with pixel size of 55 µm x 55 µm. The columns have a diameter of 10 µm,

with a depth of 250 µm in a 285 µm thick substrate. The devices collect holes

with an n-type substrate and p-doped columns connected to the readout

electronics. Using electrical characterisation lateral depletion was measured

at 2 V and full depletion at 10 V. The devices was operated over depleted

at 20 V, where it had a leakage current of 3.8 mA at room temperature.

The 3D detector assembly was placed on a precision x-y-theta stage at

the centre of a beam telescope made from 6 TimePix planar silicon detector

assemblies. The telescope has a pointing resolution for reconstructed tracks

at the device under test of 2.3 ± 0.1 µm.24

The energy collected in the pixel unit cell was reconstructed as a func-

tion of incident position of the pion beam, as shown in Fig. 2. Full charge

collection was observed for regions distant from the columns. Areas at the


6

edge of the pixel cell exhibited charge sharing, manifested as a low energy

peak in the pulse height spectra, see Fig. 2d. When the energy deposited

in the neighbouring pixel is combined into a cluster, the cluster shows full

charge collection all the way to the edge of the pixel cell, see Fig. 2e. The

centre of the pixel cell shows full charge collection from the silicon above

the column, that is 35 µm of silicon. The charge collected at the corner

of the pixel is low as the charge collected in the 35 µm of silicon above

the ohmic columns are shared between four pixels and falls below the pixel

threshold. The average detection efficiency of the full pixel was found to

Fig. 2. The pulse height spectra as a function of incident position, obtained from adouble-sided 3D detector fabricated by CNM connected to a TimePix pixel chip andilluminated by a high energy pion beam. The histograms are of the ToT counts in the

central electrode region (a), away from the central electrode and pixel edges (b). Pixelmaps showing the mean energy deposition across the pixel matrix, for a single pixel (c)and the energy in clusters (f). (d) and (e) show the histograms of the energy deposited

at the pixel edges for the single pixel and the clusters.

be 93% for normal incidence due to the low efficiency in the corners of the

pixel unit cell. As the angle of incidence was increase to 10 degrees the

detection efficiency increased to 99.8 ± 0.5% across the pixel matrix. At

an angle of 10 degrees the track traverses a full pixel within the thickness

of the sensors. This is also the angle at which the best spatial resolution

of 9.18 ± 0.1µm24 is obtained. At normal incidence the spatial resolution

is 15.8 ± 0.1µm which is in agreement with the binary resolution of the

pixel of pitch 55 µm. This is due to the low charge sharing present in the

3D detector structure. More details of the test beam analysis can be found

in.25

6. Charge collection in 3D detectors

Short 285 µm thick 3D double-sided strip sensors have been fabricated to

study charge collection after heavy irradiation. The short strip device has


7

an 80 µm pitch between columns of the same type, strips of pitch 80 µm and

4 mm strip length. They were irradiated up to a fluence of 2 x 1016 cm−2

1 MeV neq. They were tested with a 90-Sr source and readout with the

LHC speed analogue front-end Beetle chip26 integrated into the Alibava

data acquisition system.27 The detectors were cooled to -13.4 ◦C. For a full

description see.28

The results of the charge collections as a function of fluence for the de-

vice biased to 150 V in shown in Fig. 3. The 3D detectors give full charge

collection up to a fluence of 1015 cm−2 1 MeV neq. The collected charge

falls to 47% of the expected charge deposition after a fluence of 1016 cm−2

1 MeV neq. The noise remains constant as a function of fluence and there-

fore the signal-to-noise ratio as a function of fluence follows the collected

charge curve. The collected charge in the 3D detector is well modelled by

TCAD simulation without any high field effects being required. The col-

lected charge is more than that observed in a 300 µm thick planar device

operated at a bias voltage of 1000 V, which collected 30% of the expected

deposited charge after a fluence of 1016 cm−2 1 MeV neq. The larger col-

lected charge in the 3D detector is due to the higher electric fields inside

the 3D sensor and the shorter collection distances compared to the planar

devices.

0 2 4 6 8 10 120

5000

10000

15000

20000

25000

30000

Col

lect

ed c

harg

e (e

lect

rons

)

Fluence (1015 1 MeV neq cm-2)

Fig. 3. The collected charge as a func-tion of fluence for a 285 µm thick 3Dsensor operated at 150 V (open circles)and a 300 µm thick planar sensor oper-ated at 1000 V (closed diamonds).

0 5 10 15 200

5000

10000

15000

20000

25000

30000

35000

0

10

20

30

40

50

Col

lect

ed C

harg

e (e

lect

rons

)

Fluence / 1015 (1MeV neq cm-2)

Sig

nal t

o no

ise

ratio

Fig. 4. The collected charge as a func-tion of fluence for the 3D sensor oper-ated at 250–350 V (open circles) and theplanar sensor operated at 1000 V (opensquares). The 3D sensor’s S/N is shown(closed diamonds).

The collected charge in the 3D detector increases greatly with a increase

in bias voltage.28 This is shown in Fig. 4 for bias voltages up to 350 V.

The increase in charge is due to charge multiplication, which is evident for


8

devices irradiated to less than 2 x 1015 cm−2 1 MeV neq. The multiplication

is believed to be due to impact ionisation which take place in the high field

regions around the electrodes which extend through the device thickness.

At these higher bias voltages a charge of 50% of that expected is collected

after a fluence of 2 x 1016 cm−2 1 MeV neq. The charge multiplication also

leads to increased noise at the highest bias voltages and therefore lower

S/N than expected, see Fig. 4.

The Sr-90 measurements described above probe the average response

of the detector, allowing the absolute charge collection efficiency and sig-

nal to noise ratios to be extracted. To probe the relative charge collection

efficiency as a function of position in a unit cell of the device a focused

laser system is employed. A 4 µm diameter focused infrared laser spot is

raster scanned in 2 µm steps across the front surface of the sensor.29 As

the absorption depth of the 974 nm wavelength light in silicon is about

100 µm the charge collection process in the upper portion of the detector

is investigated preferentially.

Laser scans on un-irradiated sensors show a uniform collection of charge

outside of the electrode regions. In contrast to this large areas of non-

uniformity can be seen in the unit pixel of irradiated sensors. Fig. 5 shows

the charge collected when the laser is scanned across a part of the unit cell

illustrated. The sensor was irradiated to 2 x 1015 cm−2 1 MeV neq) and

biased at 260 V. Less charge is collected in regions of low field between

columns of the same doping type and an enhanced signal in the region

of high field between columns of opposite doping is observed. A higher

field region will result in faster charge collection and less time for charge

trapping resulting in greater charge collection. The relative signal from

these regions was investigated for increasing bias voltage. The shape of

the signal as a function of bias voltage was the same for all regions with a

difference between the high and low field regions of about 30%. Below 150V

an increasing voltage leads to an increase in the volume of depletion region

throughout the device and therefore an increase in the charge collected in

both regions. However, the increase in charge collection is modest. When

the voltage is increased from 150 V to 260 V the charge collection increases

dramatically to be over twice that collected at 150 V. This is due to charge

multiplication effects as mentioned above.

7. Conclusions

3D silicon detectors are now well developed. The full 3D and the double-

sided 3D detector can both be fabricated in industry. The double-sided


9

Fig. 5. Left: Diagram of the 3D strip detector under test with the scanned area showedas a dashed square. Three areas of differing electric field magnitude are labelled. Right:The response of the 3D detector biased to 260 V to collimated laser light after a fluence

of 2 x 1015 cm−2 1 MeV neq.

device to date has a better yield due to the less demanding processing tech-

nology. The double-sided 3D detector has been chosen as a sensor solution

for the first upgrade to the ATLAS silicon system; namely the IBL.

Precision scans of the pixel have been preformed in high energy test

beams. The charge deposition in a unit cell has been mapped and full

charge collection is observed in the majority of the cell outwith the elec-

trode columns. High detection efficiency across the 3D pixel matrix at zero

degree beam incident angle has been observed. This increases to 100% at

an angle of incidence of 10 degrees where charge is never deposited only

in the electrodes. The charge sharing has been shown to be less than for a

planar device resulting in lower spatial resolution (consistent with binary

readout) however, this might result in more charge being collected in the

hit pixel after irradiation.

90-Sr source test shows higher charge collected in the 3D detector com-

pered to the planar device. After an irradiation dose of 1016 cm−2 1 MeV neq

a 3D detector operated at a modest bias voltage of 150 V collected charge of

47% of the deposited charge compared to 30% in the planar device operated

at 1000 V. At higher bias voltages, (typically 300 V), charge multiplication

was observed in the 3D irradiated device. Spatially resolved laser scanning

showed a uniform response across the unit cell outside the column region


10

for an un-irradiated sensor when operated above full depletion. After heavy

irradiation the response was non-uniform with an area of enhanced signal

in the region of high field between columns of opposite doping.

References

1. M. Capeans, G. Darbo, K. Einsweiller, M. Elsing, T. Flick, M. Garcia-Sciveres, C. Gemme, H. Pernegger, O. Rohne and R. Vuillermet, ATLAS

Insertable B-Layer Technical Design Report, CERN technical report CERN-LHCC-2010-013. ATLAS-TDR-019, CERN (Geneva, Switzerland, 2010).

2. S. I. Parker, C. J. Kenney and J. Segal, Nucl. Instr. and Meth A 395, 328(1997).

3. G. Kramberger, V. Cindro, I. Mandic, M. Mikuz and M. Zavrtanik, Nucl.

Instr. and Meth A 476, 645 (2002).4. A. G. Bates and M. Moll, Nucl. Instr. and Meth A 555, 113 (2005).5. S. Parker, C. Kenney, J. Segal and C. Storment, IEEE Trans. Nucl. Sci. 46,

1224 (1999).6. G. Pellegrini, M. Lozano, M. Ulln, R. Bates, C. Fleta and D. Pennicard, Nucl.

Instr. and Meth A 592, 38 (2008).7. C. DaVia and S. J. Watts, Nucl. Instr. and Meth A 603, 319 (2009).8. C. J. Kenney, S. Parker and E. Walckiers, IEEE Trans. Nucl. Sci. 48, 2405

(2001).9. T.-E. Hansen, A. Kok, T. A. Hansen, N. Lietaer, M. Mielnik, P. Storas,

C. DaVia, J. Hasi, C. Kenney and S. Parker, JINST 4, p. P03010 (2009).10. A. Kok, T. E. Hansen, T. A. Hansen, N. Lietaer, S. Anand, C. Kenney,

J. Hasi, C. DaVia and S. I. Parker, Fabrication of 3d silicon sensors, inProceedings of Science, the 19th International Workshop on Vertex Detectors,(Loch Lomand, UK, 2010).

11. I. Perica, L. Blanquart, G. Comes, P. Denes, K. Einsweiler, P. Fischer,E. Mandelli and G. Meddeler, Nucl. Instr. and Meth A 565, 178 (2006).

12. A. Kok, Fabrication of full 3d active edge sensors(March 2011), 6th “Trento”workshop on advanced silicon radiation detectors. https://indico.cern.

ch/contributionListDisplay.py?confId=114255.13. J. Hasi, Status of 3d sensors processing at stanford(February 2010), 5th

“Trento” workshop on advanced silicon radiation detectors. http://agenda.hep.manchester.ac.uk/conferenceDisplay.py?confId=1181.

14. A. Zoboli, M. Boscardin, L. Bosisio, G.-F. Betta, S. Piemonte, C. Ronchinand N. Zorzi, IEEE Trans. Nucl. Sci. 55, 2775 (2008).

15. D. Pennicard, G. Pellegrini, M. Lozano, R. Bates, C. Parkes, V. O’Shea andV. Wright, IEEE Trans. Nucl. Sci. 54, 1435 (2007).

16. G. D. Betta, A. Bagolini, M. Boscardin, L. Bosisio, P. Gabos, G. Giacomini,C. Piemonte, M. Povoli, E. Vianello and N. Zorzi, Development of modified3d detectors at fbk, in IEEE Nuclear Science Symposium Conference Record ,(Knoxville, USA, 2010).

17. O. Koybasi, D. Bortoletto, T. Hansen, A. Kok, T. Hansen, N. Lietaer, G. U.


11

Jensen, A. Summanwar, G. Bolla and S. W. L. Kwan, IEEE Trans. Nucl.

Sci. 57, 2897 (2010).18. O. Koybasi, E. Alagoz, A. Krzywda, K. Arndt, G. Bolla, D. Bortoletto, T.-E.

Hansen, T. A. Hansen, G. U. Jensen, A. Kok, S. Kwan, N. Lietaer, R. Rivera,I. Shipsey, L. Uplegger and C. DaVia, IEEE Trans. Nucl. Sci. 58, 1315 (2011).

19. M. Garcia-Sciveres, D. Arutinov, M. Barbero, R. Beccherle, S. Dube,D. Elledge, J. Fleury, D. Fougeron, F. Gensolen, D. Gnani, V. Gromov,T. Hemperek, M. Karagounis, R. Kluit, A. Kruth, A. Mekkaoui, M. Menouniand J.-D. Schipper, Nucl. Instr. and Meth A 636, 155 (2011).

20. P. Grenier, G. Alimonti, M. Barbero, R. Bates, E. Bolle, M. Borri,M. Boscardin, C. Buttar, M. Capua, M. Cavalli-Sforza, M. Cobal, A. Cristo-foli, G.-F. D. Betta, G. Darbo, C. D. Vi, E. Devetak, B. DeWilde, B. D. Giro-lamo, D. Dobos, K. Einsweiler, D. Esseni, S. Fazio, C. Fleta, J. Freestone,C. Gallrapp, M. Garcia-Sciveres, G. Gariano, C. Gemme, M.-P. Giordani,H. Gjersdal, S. Grinstein, T. Hansen, T.-E. Hansen, P. Hansson, J. Hasi,K. Helle, M. Hoeferkamp, F. Hgging, P. Jackson, K. Jakobs, J. Kalliopuska,M. Karagounis, C. Kenney, M. Khler, M. Kocian, A. Kok, S. Kolya, I. Ko-rokolov, V. Kostyukhin, H. Krger, A. L. Rosa, C. Lai, N. Lietaer, M. Lozano,A. Mastroberardino, A. Micelli, C. Nellist, A. Oja, V. Oshea, C. Padilla,P. Palestri, S. Parker, U. Parzefall, J. Pater, G. Pellegrini, H. Pernegger,C. Piemonte, S. Pospisil, M. Povoli, S. Roe, O. Rohne, S. Ronchin, A. Rovani,E. Ruscino, H. Sandaker, S. Seidel, L. Selmi, D. Silverstein, K. Sjbk, T. Slav-icek, S. Stapnes, B. Stugu, J. Stupak, D. Su, G. Susinno, R. Thompson, J.-W.Tsung, D. Tsybychev, S. Watts, N. Wermes, C. Young and N. Zorzi, Nucl.

Instr. and Meth A 638, 33 (2011).21. A. Micelli, K. Helle, H. Sandaker, B. Stugu, M. Barbero, F. Hgging,

M. Karagounis, V. Kostyukhin, H. Krger, J.-W. Tsung, N. Wermes, M. Ca-pua, S. Fazio, A. Mastroberardino, G. Susinno, C. Gallrapp, B. D. Giro-lamo, D. Dobos, A. L. Rosa, H. Pernegger, S. Roe, T. Slavicek, S. Pospisil,K. Jakobs, M. Khler, U. Parzefall, G. Darbo, G. Gariano, C. Gemme,A. Rovani, E. Ruscino, C. Butter, R. Bates, V. Oshea, S. Parker, M. Cavalli-Sforza, S. Grinstein, I. Korokolov, C. Pradilla, K. Einsweiler, M. Garcia-Sciveres, M. Borri, C. D. Vi, J. Freestone, S. Kolya, C. Lai, C. Nellist, J. Pa-ter, R. Thompson, S. Watts, M. Hoeferkamp, S. Seidel, E. Bolle, H. Gjersdal,K.-N. Sjoebaek, S. Stapnes, O. Rohne, D. Su, C. Young, P. Hansson, P. Gre-nier, J. Hasi, C. Kenney, M. Kocian, P. Jackson, D. Silverstein, H. Dav-etak, B. DeWilde, D. Tsybychev, G.-F. D. Betta, P. Gabos, M. Povoli,M. Cobal, M.-P. Giordani, L. Selmi, A. Cristofoli, D. Esseni, P. Palestri,C. Fleta, M. Lozano, G. Pellegrini, M. Boscardin, A. Bagolini, C. Piemonte,S. Ronchin, N. Zorzi, T.-E. Hansen, T. Hansen, A. Kok, N. Lietaer, J. Kallio-puska and A. Oja, Nucl. Instr. and Meth A 650, 150 (2011).

22. J. Weingarten, Irradiation and beam tests qualification for atlas ibl pixelmodules(September 2011), The 9th international conference on position sen-sitive detectors. http://indico.cern.ch/conferenceDisplay.py?confId=

48618.23. X. Llopart, R. Ballabriga, M. Campbell, L. Tlustos and W. Wong, Nucl.


12

Instr. and Meth A 581, 485 (2007).24. K. Akiba, M. Artuso, R. Badman, A. Borgia, R. Bates, F. Bayer, M. van

Beuzekom, J. Buytaert, E. Cabruja, M. Campbell, P. Collins, M. Crossley,R. Dumps, L. Eklund, D. Esperante, C. Fleta, A. Gallas, M. Gandelman,J. Garofoli, M. Gersabeck, V. V. Gligorov, H. Gordon, E. H. Heijne, V. Hei-jne, D. Hynds, M. John, A. Leflat, L. F. Llin, X. Llopart, M. Lozano, D. Ma-neuski, T. Michel, M. Nicol, M. Needham, C. Parkes, G. Pellegrini, R. Plack-ett, T. Poikela, E. Rodrigues, G. Stewart, J. Wang and Z. Xing, Nucl. Instr.

and Meth A PP (2011).25. A. M. Raighne, K. Akiba, L. Alianelli, R. Bates, M. van Beuzekom, J. Buy-

taert, M. Campbell, P. Collins, M. Crossley, R. Dumps, L. Eklund, C. Fleta,A. Gallas, M. Gersabeck, E. N. Gimenez, V. V. Gligorov, M. John, X. Llopart,M. Lozano, D. Maneuski, J. Marchal, M. Nicol, R. Plackett, C. Parkes, G. Pel-legrini, D. Pennicard, E. Rodrigues, G. Stewart, K. J. S. Sawhney, N. Tartoniand L. Tlustos, JINST 6, p. P05002 (2011).

26. S. Lochner and M. Schmelling, The Beetle Reference Manual - chip version

1.3, 1.4 and 1.5, CERN technical report LHCb-2005-105. CERN-LHCb-2005-105, CERN (Geneva, Switzerland, 2006).

27. R. Marco-Hernandez, IEEE Trans. Nucl. Sci. 56, 1642 (2009).28. R. L. Bates, C. Parkes, B. Rakotomiaramanana, C. Fleta, G. Pellegrini,

M. Lozano, J. P. Balbuena, U. Parzefall, M. Koehler, M. Breindl and X. Blot,Nuclear Science, IEEE Transactions on PP(December 2011).

29. A. Zoboli, G.-F. D. Betta, M. Boscardin, L. Bosisio, S. Eckert, S. Khn,U. Parzefall, C. Piemonte, S. Ronchin and N. Zorzi, Nucl. Instr. and Meth A

604, 238 (2009), ¡ce:title¿PSD8¡/ce:title¿ ¡ce:subtitle¿Proceedings of the 8thInternational Conference on Position Sensitive Detectors¡/ce:subtitle¿.

Documents

3D silicon detectors 1. Introduction The LHC saw a change in the